DI

SC

US

SI

ON

P

AP

ER

S

ER

IE

S

Forschungsinstitut zur Zukunft der ArbeitInstitute for the Study of Labor

Labor Mobility in an Enlarged European Union

IZA DP No. 6485

April 2012

Martin Kahanec

Labor Mobility in an

Enlarged European Union

Martin Kahanec Central European University,

IZA and CELSI

Discussion Paper No. 6485 April 2012

IZA

P.O. Box 7240 53072 Bonn

Germany

Phone: +49-228-3894-0 Fax: +49-228-3894-180

E-mail: iza@iza.org

Any opinions expressed here are those of the author(s) and not those of IZA. Research published in this series may include views on policy, but the institute itself takes no institutional policy positions. The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center and a place of communication between science, politics and business. IZA is an independent nonprofit organization supported by Deutsche Post Foundation. The center is associated with the University of Bonn and offers a stimulating research environment through its international network, workshops and conferences, data service, project support, research visits and doctoral program. IZA engages in (i) original and internationally competitive research in all fields of labor economics, (ii) development of policy concepts, and (iii) dissemination of research results and concepts to the interested public. IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper should account for its provisional character. A revised version may be available directly from the author.

IZA Discussion Paper No. 6485 April 2012

ABSTRACT

Labor Mobility in an Enlarged European Union* The 2004 and 2007 enlargements of the EU extended the freedom of movement to workers from the twelve new member states mainly from Central Eastern Europe. This study summarizes and comparatively evaluates what we know about mobility in an enlarged Europe to date. The pre-enlargement fears of free labor mobility proved to be unjustified. No significant detrimental effects on the receiving countries’ labor markets have been documented, nor has there been any discernible welfare shopping. Rather, there appear to have been positive effects on EU’s productivity. The sending countries face some risks of losing their young and skilled labor force, but free labor mobility has relieved them of some redundant labor and the associated fiscal burden. They have also profited from remittances. Of key importance for the sending countries is to reap the benefits from brain gain and brain circulation in an enlarged EU. For the migrants the benefits in terms of better career prospects have with little doubt exceeded any pecuniary and non-pecuniary costs of migration. In conclusion, the freedom of movement in the EU provides for a triple-win situation for the receiving and sending countries as well as for migrants themselves, provided the risks are contained and efficient brain circulation is achieved. JEL Classification: F22, J61 Keywords: EU labor markets, migration, EU enlargement, labor mobility, free movement

of workers, transitional arrangements, new member states, European Union Corresponding author: Martin Kahanec Department of Public Policy Central European University Nádor u. 11 H-1051 Budapest Hungary E-mail: KahanecM@ceu.hu

* Forthcoming in “International Handbook of the Economics of Migration”. The author thanks Amelie F. Constant, Klaus F. Zimmermann and an anonymous referee for providing a number of suggestions that helped to improve the chapter significantly. I remain responsible for any mistakes still present.

2

Daughter: “What’s over there, Mom?”

Mother: “There is nothing there – there is the East Bloc.”

Conversation between a mother and her daughter pointing in the direction of Slovakia on

a hilltop in Austria near the Czechoslovak border sometime in 1987.

1. Introduction

The freedom of movement of workers is one of the four fundamental pillars of economic

integration in the European Union (EU), which also includes the free mobility of capital,

goods, and services. A central objective of free mobility is to enable EU citizens to seek

employment, and any social benefits attached with it, in any of the EU member states.

From the economic perspective free labor mobility improves the allocative efficiency of

EU labor markets, thus buttressing the EU’s economy and alleviating some of its

demographic challenges (Zimmermann, 2005; Kahanec and Zimmermann, 2010). Yet,

with the process of EU enlargement expanding this freedom to new member states, free

labor mobility constitutes one of the most sensitive, and often challenged, freedoms in the

EU.

The controversies surrounding the freedom of movement of labor culminated when

Cyprus, the Czech Republic, Estonia, Latvia, Lithuania, Malta, Hungary, Poland,

Slovakia, and Slovenia, referred to as the EU10, joined the EU in 2004 and carried on in

2007 when Bulgaria and Romania, the EU2, followed suit.1 These controversies were

1 The respective abbreviations used in the text below are: CY, CZ, EE, LV, LT, MT, HU, PL, SK, SI, BG, and RO. EU8 denotes EU10 minus Cyprus and Malta; EU8+2 includes EU8 and EU2. EU15 includes

3

probably rooted in the history of deep political, economic and social separation during the

Cold War. This separation had severely limited mobility and contact across the East-West

limits and resulted in a fissure in the European identity along the Iron Curtain.

Presumably the economic disparities between the new and old member states, combined

with the large scale of these enlargements, created grounds for a widespread perception in

the EU15 of EU8+2 migrants as a threat to their labor markets and welfare systems, and

explain the magnitude of such controversies at least partly.2 As a consequence, a policy

instrument – transitional arrangements – was adopted allowing member states to keep

their labor markets closed for citizens from new member states for up to 7 years, with

revisions required after 2 and 5 years, following their accession.3

In the EU8+2 free mobility was seen as a way out of the difficulties stemming from labor

market mismatches and excess labor inherited from the process of their difficult post-

socialist transformation. In spite of some fears of brain drain, overall, the expectations of

faster convergence to the living standards of the old member states following their

Austria (abbreviated AT), Belgium (BE), Denmark (DK), Finland (FI), France (FR), Germany (DE), Greece (EL), Ireland (IE), Italy (IT), Luxembourg (LU), the Netherlands (NL), Portugal (PT), Spain (ES), Sweden (SE) and the United Kingdom (UK). 2 Some of the early forecasts added to the fears in the EU15 by predicting rather high east-west migration flows (Sinn et al., 2000), possibly even undermining the welfare state in the receiving countries (Sinn and Ochel, 2003). More moderate migration rates in the vicinity of actual post-enlargement migration flows were predicted by e.g. Layard et al. (1992), Bauer and Zimmermann (1999), Dustmann et al. (2003), IOM (1998); see also Zaiceva and Zimmermann, (2008) and Brücker et al. (2009). See Canoy et al. (2010) for a thorough account of the links between public perception, migrants’ labor market outcomes, and migration policies. 3 Cyprus and Malta were exempt from such restrictions. Ireland, the UK, and Sweden opened up their labor markets immediately following the 2004 accession, while Germany and Austria imposed restrictions up until the end of the seven-year period, albeit simplifying some of the procedures. The other old member states had gradually opened up by May 1, 2009. As for the 2007 enlargement, ten member states opened up their labor markets during the first 2-year phase: the Czech Republic, Estonia, Cyprus, Latvia, Lithuania, Poland, Slovenia, Slovakia, Finland and Sweden. By the end of the second phase on January 1, 2012, Denmark, Greece, Hungary, Portugal, Slovenia and Spain opened up as well, with Austria, Germany, Belgium, France, Ireland, Luxembourg, Malta, the Netherlands, and the UK still applying transitional arrangements as of January 2012. In July 2011 the EC authorized Spain to reinstate restrictions for Romanian workers until the end of 2012.

4

accession framed the expectations in the acceding countries quite favorably towards this

process.

This chapter reviews what we know about labor mobility in the EU following the two

recent waves of enlargement in 2004 and 2007. We in particular evaluate the experience

with post-enlargement migration in an enlarged EU in view of the fears and hopes

attached to it in the sending and receiving countries, and by the migrants themselves. The

next section provides a theoretical account of possible effects of free mobility in sending

and receiving labor markets. We then describe how enlargement affected labor mobility

in the EU, and what measurable effects can be documented empirically. In the subsequent

section we shed some light on what migration flows can be expected in the near future.

Finally, we discuss a number of lessons that can be learned, and conclude.

2. A theoretical account

To pinpoint the diverse social, political, or economic factors behind the reserved attitudes

towards free labor mobility in the old member states and the more relaxed perceptions in

the new member states would take a major study on its own. We undertake a narrower

question here: does economic theory predict any worrisome effects of increased mobility

for the receiving and sending countries? And does it in fact predict any significant

migration flows in an enlarged EU?

5

The answer to the latter question is in all likelihood “yes”. Harris and Todaro (1970)

point to the significance of (expected) regional disparities in the standard of living for the

migration decision. More generally, international disparities in the levels (and

distribution) of earnings and income, net of migration costs, chances to pursue a

rewarding career and avoid unemployment, the cost of living, or the availability and

quality of public goods and amenities are proposed in the literature as key drivers of

migration (Massey, 1990; Borjas, 1999; Bonin et al., 2008).4 Others, such as the

generosity of the welfare system are more controversial (De Giorgi and Pellizzari, 2006;

Giulietti et al., 2012). Stark (1991) advanced the view that for the household as a

decision-making unit it may be worthwhile to have one or more of its members abroad as

a strategy of risk sharing.

Factors such as those listed above may affect various subpopulations differently. The

costs of migration and adjustment in the host economy, pecuniary and non-pecuniary,

depend on the geographical, linguistic, and cultural distances between (subpopulations in)

the sending and receiving countries (Chiswick and Miller, 2011). The human capital

theory predicts that the migration decision also depends on age and skills of potential

migrants, as these determine their capacity to adjust in the host country and thus benefit

from migration (Becker, 1957; Sjaastad, 1962). As a result, people who decide to migrate

and stay in the receiving country may be positively or negatively self-selected based on

their observable or unobservable characteristics (Borjas, 1987; Chiswick, 1999).

4 Besides these economic factors, family, ethnic, or social ties; natural catastrophes; social and political crises; as well as discrimination or persecution may result in significant movement of people (Mincer, 1978; Massey, 1990).

6

Based on these arguments and given the initial disparities in many socio-economic

variables, it is probably just to say that the expectations of non-negligible migration rates

between new and old member states following the liberalization of the new EU citizens’

access to old member states’ labor markets were justified. Given the linguistic, cultural,

institutional, and socio-economic diversity in Europe one could also well expect uneven

migration rates across source and host populations. But are there any theoretical reasons

to justify the pre-enlargement fears of various negative economic effects caused by such

migration flows?

The impact of migration on sending and receiving labor markets can be pinned down in a

simple economic model drawing on the idea that the effects of immigration depend on the

degree of substitutability or complementarity of migrant and non-migrant (native or

staying) labor (Chiswick, Chiswick, and Karras, 1992; Chiswick, 1980, 1998). It is useful

to consider the effects of inflow and outflow of skilled and unskilled workers in the

respective labor markets. In Figure 1 we look at the effects in a receiving country.

Immigration increases the supply of high-skilled workers from 0H to 1H in panel a.

This drives the equilibrium and the corresponding wage level from 0A to 1A , along the

original demand curve 0hD . The increased employment of high-skilled workers, through

complementarity between high- and low-skilled labor, increases the demand for low-

skilled workers from 0lD to 1

lD in panel b. Assuming a competitive low-skilled market,

given the supply of low-skilled workers at 0L , high-skilled immigration leads to as shift

of the equilibrium from 0B to 1B and the corresponding wage increase. If, however, a

(binding) minimum wage is set at Ulw , high skilled immigration reduces low-skilled

7

unemployment from 00ULL − to 10

ULL − . Through complementarity of low- and high-

skilled workers the corresponding increase in low-skilled employment feeds back into the

high-skilled market and so mitigates the initial decrease of equilibrium wage, increasing

it from 1hw to 2

hw .

Low-skilled immigration, on the other hand, decreases the wage of low skilled workers

from 0lw to 2

lw along the initial demand curve under a competitive market, or it increases

unemployment from 00ULL − to 01

ULL − under a wage floor. In the former case the

increase in low-skilled employment increases the demand for high-skilled workers lifting

the high-skilled wage from 0hw to 3

hw .

[Figure 1 about here]

In a sending country, correspondingly, high-skilled out-migration increases high-skilled

workers’ wages by moving the equilibrium from 0M to 1M . Lower high-skilled

employment leads to a lower demand for low-skilled workers, either reducing their wage

from 0lw to 1

lw or increasing low skilled unemployment from 00ULL − to 10

ULL − under a

binding minimum wage. In the latter case this feeds back into the high-skilled market

through complementarity of high- and low-skilled labor, thus shifting the demand for

high skilled labor down, decreasing their wage from 1hw to 2

hw . Low-skilled out-

migration, on the other hand, increases low-skilled competitive market wage from 0lw to

2lw , or decreases unemployment from 00

ULL − to 01ULL − , under the minimum wage

8

regime. By the complementarity argument, in the former case the lower low-skilled

employment decreases demand for high-skilled labor and thus high-skilled wage from

0hw to 3

hw .

[Figure 2 about here]

This straightforward analysis elucidates the redistributive consequences of immigration

and out-migration. Clearly these depend on whether migration concerns low-skilled or

high-skilled workers. Consider, for example, the case of high-skilled post-enlargement

migration. The winners of enlargement would then be low-skilled workers in the

receiving countries benefiting from higher wages or lower unemployment. In the sending

countries, the staying high-skilled workers could also benefit unless the weakened

demand for low-skilled workers resulted in lower low-skilled employment and, as a

consequence, lower productivity of high-skilled workers in spite of their increased

scarcity. High-skilled workers in the receiving countries could be among the losers of

enlargement, but not if the increased demand for low-skilled labor resulted in their higher

employment and thus an increased productivity of high-skilled workers in spite of their

increased relative abundance. Low-skilled workers in the sending countries would clearly

lose either in terms of higher unemployment or lower wages. One can similarly track the

redistributive effects of low-skilled migration in this model.5

5 See Kahanec and Zimmermann (2009) for more on redistributive effects of migration.

9

While this textbook model elucidates some potential redistributive effects of post-

enlargement migration, many other important factors may considerably change or even

reverse some of its predictions. For example, economic migration can be expected to

improve the allocation of labor and human capital. Moreover, migration proliferates

cross-regional and cross-border social ties, thus acting as a vehicle for international flows

of goods and services, capital, as well as ideas and knowledge (Bonin et al., 2008). The

resulting improved productivity may benefit all types of labor in sending as well as

receiving countries. Further economic benefits may result from increased ethnic diversity

in receiving countries (Ottaviano and Peri, 2006)

However, a range of psychological, linguistic, institutional, or legislative barriers, as well

as discrimination, may impede immigrant adjustment in the host society, thus hindering

some of the positive effects migration may entail (Constant, Kahanec and Zimmermann,

2009; Kahanec, Kim and Zimmermann, 2011). Such barriers may for example result in

weaker labor market outcomes and, as a consequence, an increase in migrants’ demand

for welfare (Borjas, 1999; Brücker et al., 2002; Kahanec, Kim and Zimmermann, 2011;

Zimmermann et al., 2012). Ethnic identity is another factor that may positively or

negatively affect adjustment in host labor markets (Constant and Zimmermann,

forthcoming).

Whether given migration flows should be considered as high- or low-skilled depends on

the migrant’s skill level relative to the skill level in sending and receiving countries. The

same migrant can thus be seen as skilled from the perspective of the sending country but

10

unskilled in the receiving country. A similar discrepancy may also arise if skills are not

perfectly transferable from sending to receiving countries; the speed of adjustment then

determines the effective skill level of migrants in host countries.6 Downskilling, whereby

immigrant workers’ potential is underutilized in jobs below their skill level, has been

documented also in the context of post-enlargement migration (Kahanec and

Zimmermann, 2010; Kureková, 2011; Hazans 2012).7 Such considerations point out the

importance of skill measurement when evaluating the effects of migration flows.

This theoretical account of migration illustrates that the scale and skill composition of

post-enlargement migration is particularly important for the evaluation of its effects in

sending and receiving economies. For the receiving countries the degree and speed of

adjustment of immigrants is another important variable. It also shows that without strong

assumptions the effects of enlargement are hard to evaluate unequivocally based on

theoretical grounds. Similarly, any a priori fears of enlargement are hard to justify

theoretically. To evaluate the scale, composition, and effects of post-enlargement

migration one needs to look at hard data.

3. The scale and composition of post-enlargement migration

The gradual opening-up of western European labor markets instigated by the 2004 and

2007 EU enlargements enabled many workers from the new member states to seek

6 See Kahanec and Zimmermann (2009). 7 Whereas the formal recognition of qualification obtained within the EU in another EU member state has been significantly simplified by EU legislation, informational asymmetries, linguistic and other barriers still obstruct the adjustment of within-EU migrants.

11

employment and pursue careers in the more prosperous part of an enlarged EU. This has

dramatically changed the migration landscape in Europe and led to substantial east-west

migration flows.

Given the scarcity of migration data, to evaluate the scale and composition of post-

enlargement migration is a formidable task. We therefore look at various data sources and

the available literature to triangulate some of the most important trends. According to the

data provided in Holland et al. (2011), in 2004 there were about a million citizens from

the EU8, and almost another million EU2 nationals, residing in the EU15.8 By 2009, just

five years later, the total number of EU8 and EU2 citizens residing in the EU15 increased

by about 150% and reached almost five million (Table 1). In effect, the combined

populations of citizens from EU8 and EU2 countries residing in the EU15 constituted

1.22% of the total EU15 population and 4.75% of combined populations of EU8 and EU2

countries.9

[Table 1 about here]

8 While this dataset provides probably the most comprehensive account of migration flows between the new and old member states known to us, it has to be acknowledged that a number of issues arise with it. These mainly arise because of the lack of adequate infrastructure to collect data enabling us to measure migration flows in the EU. For example, data is often based on population statistics by citizenship, and changes in respective stocks are interpreted as migration flows. Deaths and births, legalizations, as well as citizenship acquisition, are included in these flows, although they should not be interpreted as migration. Latvia and Estonia are especially problematic in this respect, as these countries host large populations of non-citizens, who are treated in various destination countries in different ways. Data from Ireland and the UK are similarly problematic, as they are based on interpolations from the respective labor force surveys rather than large-scale administrative or census data, which may have large error especially for evaluating the sizes of populations originating from smaller source countries. Looking at foreign-born populations does not help to solve all these issues; for example, many migrants from the Baltic states were born in other republics of the Soviet Union. Various registers have their own problems, as migrants often fail to deregister. The statistics that we discuss below may therefore over- or under-represent true migration flows and need to be interpreted with these caveats in mind. 9 For 2007 these figures are slightly higher than those reported by Brücker and Damelang (2009) or Brücker et al. (2009), and in the range of those provided by European Commission (2008a, b).

12

Whereas over the five-year period preceding 2004 the average annual inflow to the EU15

from the EU8 was about 58,000, in the five years after 2004 this has risen to 256,000

annually. The corresponding figures for the EU2 were 129,000 and 330,000, respectively

(Table 1). The dynamics of these flows are visualized in Figure 3. We observe an

increasing dynamic of inflows from new to old member states until 2007, followed by a

significant slow down during the financial crisis in 2008 and 2009. EU8 citizens reacted

to enlargement with some delay, with peak migration attained only in 2006 and 2007, two

years after their accession. The response of EU2 citizens was considerably swifter and

more pronounced, reaching peak migration flows already in the year of enlargement.10

The slow down of 2008 and 2009 indicates that the worsened economic prospects in

some of the receiving countries may have discouraged potential migrants.

[Figure 3 about here]

From where and where to did migrants from the new member states go? The most

important sending countries are Romania and Poland, which in 2009 together accounted

for three quarters of all migrants from the EU8 and EU2 in the EU15. Figure 4 shows that

the numbers of citizens from new member states in the EU15 as percentages of respective

source populations. A clear picture that emerges is that the most significant sending

countries, relative to their populations, are Romania, Lithuania, and Bulgaria. The Czech

Republic, Slovenia, and Hungary exhibit the lowest shares of their population residing in

the EU15.

10 That the 2004 accession took place on May 1, whereas in 2007 it was January 1, can at best only partly explain this difference in response.

13

[Figure 4 about here]

As for the receiving countries, in 2009 the most significant of the EU15 host countries for

EU8 citizens were Germany and the UK, jointly hosting 62% of them. For EU2 citizens

the two most significant destinations were Italy and Spain, in 2009 each hosting more

than 40% of all EU2 citizens residing in the EU15.11 In Figure 5 we distinguish countries

by the period in which they opened up their labor markets to citizens from new member

states.

[Figure 5 about here]

Panel a of Figure 5 shows that the growth in population of EU8 citizens increased

significantly, although to a different degree, in each of the countries that liberalized

access to their labor market as of May 1, 2004. Remarkably, many of the countries that

opened up their labor markets later – including Denmark, the Netherlands, Luxembourg,

Finland, and Austria – have similarly experienced an increase in the rate of growth of

their EU8 populations following the 2004 enlargement. A possible explanation is that EU

accession removed some bureaucratic and psychological barriers to moving to old

member states or that EU8 citizens circumvented labor market barriers mainly by coming

as self-employers.12 In general, Figure 5 documents that populations of EU8 citizens

11 The size of these populations needs also to be interpreted in the context of total immigrant populations, as people originating from EU10 or EU2 constitute only a smaller fraction of all immigrants in EU15 (Kahanec, Zaiceva and Zimmermann, 2010). 12 Even in countries applying transitional arrangements restrictions for EU8+2 migrants were relaxed upon their countries’ EU accession. This includes preferential treatment in access to work permits vis-à-vis third country nationals, the freedom of establishment of a business for self-employed, and the freedom to provide services and thus post workers in the EU15 (excepting Austria and Germany).

14

continued to increase across old member states throughout the studied period, the rate of

growth of these populations increased in the post-enlargement period as compared to the

pre-2004 period, and a delayed liberalization of labor market access in some EU15

countries may have diverted some migrants but did not prevent their EU8 populations

from growing.

Concerning citizens of EU2 countries, their access to most EU15 labor markets continued

to be restricted throughout the studied period. Nevertheless, EU2 populations increased

significantly in southern Europe, most notably in Spain, Italy, but also in Greece (Figure

6). In Scandinavia, Bulgarian and Romanian populations continued to be rather small,

although since 2007 there appear to be significant growth rates in Denmark and Sweden.

Among the other EU15 countries Austria, Ireland, Luxembourg, and Belgium hosted the

most dynamic EU2 populations.

[Figure 6 about here]

The trends discussed above point at an important phenomenon that characterizes post-

enlargement migration, namely the geographic diversion of migration flows. Figure 7

demonstrates that for EU8 citizens the relative importance of the UK, Ireland but also

Spain as host countries increased substantially, while the traditional host countries,

Germany and Austria, lost their share quite dramatically. For EU2 citizens the shares of

Spain and Italy increased steeply, at the expense of mainly Germany, but also Austria and

15

France. The effects of this diversion may be long lasting due to the power of immigrant

networks for the migration decision (Delbecq and Waldorf, 2010).

[Figure 7 about here]

As concerns the skill composition of citizens from new member states residing in the

EU15 in the early post-enlargement period, a number of early studies indicate that the

majority of EU8 immigrants had medium educational attainment, and almost a quarter of

them attained high education (European Commission 2008b; Brücker and Damelang

2009; Bruecker et al. 2009). Brücker and Damelang report that in 2006 among EU8

migrants in the EU15 17% had low and 22% had high educational attainment. The

corresponding figures for EU2 migrants were 29% and 18%. Among the natives in the

EU15 there were 27% of them with low and the same percentage with high educational

attainment. Holland et al. (2011) find that Luxembourg, Demark, Sweden, and Ireland are

most popular among high-skilled workers while low-skilled workers are more likely to go

to Greece, Portugal, Spain, Belgium, Netherlands, and Finland. Furthermore, this study

finds that for most of the EU8+2 countries’ migrants heading to the EU15 over-represent

the high- (except for Estonia, Slovenia, and Lithuania) as well as low-skilled (excepting

Hungary and Latvia) domestic populations, but under-represent the medium-skilled

population.

A book edited by Kahanec and Zimmermann (2010) systematically summarizes the

available evidence on the scale, composition, and effects of free labor mobility in the

16

early post-enlargement period. Kahanec, Zaiceva, and Zimmermann (2010) provide a

broad account of post-enlargement migration in the EU documenting the cross-country

differences in the scale and composition of these flows and their effects. They in

particular argue that EU enlargement has had different effects in countries that opened up

their labor market early on, such as the UK, and those that strictly applied transitional

arrangements, such as Germany. For example, they document that while the skill

composition of EU8 immigrants improved after enlargement in the UK, it worsened in

Germany. Blanchflower and Lawton (2010) report that in the UK EU8 migrants had a

high incidence of self-employment and high employment rates, and were well skilled.

Barrett (2010) finds that the EU10 migrants in Ireland had very high employment rates

and levels of education comparable to the natives. He also finds evidence for

downskilling accompanied by relatively lower wages.

Brenke, Yuksel, and Zimmernann (2010) document that post-enlargement migrants from

the EU8 in Germany were predominantly male and young but were less educated and

older than EU8 migrants had been previously. The authors also report higher self-

employment rates but lower earnings and lower quality jobs for these immigrants. Self-

employment rates as high as 38% for post-enlargement migrants from the EU10 in

Germany and 51% for those coming from EU2 in the UK in 2007 reported by European

Commission (2008b) may signify inefficient spurious self-employment as a way to

circumvent transitional arrangements imposed in these cases. A study by de la Rica

(2010) reports that EU8+2 immigrants in Spain were predominantly young and had

secondary education, allowing them to achieve high employment rates, but they also

17

struggled with relatively high unemployment. Importantly, she also reports lack of

adjustment as concerns job quality. Gerdes and Wadensjö (2010) find that in Sweden

post-enlargement migrants were relatively young and highly educated, but their earnings

and employment rates were not as high as those of the natives. While before enlargement

immigration to Sweden from the EU10 was dominated by females, in the post-

enlargement its gender composition became much more even. A preliminary comparative

evaluation of these experiences with post-enlargement migration in Europe indicate that

transitional arrangements backfired in that they implied a negative selection of incoming

workers in terms of their skills and age (Kahanec and Zimmermann, 2010).

Kaczmarczyk, Mioduszewska, and Żylicz (2010) provide evidence on the main sending

country, Poland, arguing that the economic effects of relatively large out-migration are

moderate. They propose that post-enlargement migration may foster the process of

modernization in Poland, to the extent brain circulation facilitates restructuring and a

higher allocative efficiency. Hazans and Philips (2010) and Hazans (2012) find that in

the Baltic states’ post-enlargement migrants were significantly less educated than stayers,

with medium-skilled workers being most likely to move after accession. They do not find

evidence for brain drain but report significant brain waste in the form of downskilling.

Galgóczi, Leschke, and Watt (2012) enrich the literature by shedding light on skill-

mismatches in an enlarged EU and the role of trade unions in bridging these mismatches.

Using an innovative web-based survey WageIndicator, Tijdens and van Klaveren (2012)

document that among EU15 residents born in the EU10 only 65 percent report a correct

18

job-education match compared to 74% for the whole sample and 72% for all migrants.

Kureková (2011) stresses the importance of skill-mismatches in the sending EU10

countries, and their interaction with the welfare state for the scale and composition of

post-enlargement migration flows.

In our own analysis based on the 2009 wave of the EU Labour Force Survey we

reconstruct immigrant cohorts using the year of arrival for residents born in the EU10 and

EU2. We consider the population above and including 16 years of age, excluding

conscripts on compulsory military or community service as well as anyone whose highest

level of education or training successfully completed was attained after his or her

immigration to the current country of residence in the EU15. Figure 8 reports the shares

of EU10 and EU2 immigrants with high, medium, and low level of education.13 We

observe that with enlargement the share of EU10 migrants with high educational

attainment residing in the EU15 increased substantially.14 Interestingly, the share of

highly educated EU10 migrants increased already in 2003, which might indicate that

even the prospect of impending EU accession sealed already in 2003 attracted many

educated EU10 citizens. We also observe that during the first three years following the

accession the share of low educated EU10 migrants was lower than before accession. In

2007 and 2009 we however observe somewhat higher shares of low educated EU10

migrants. This is consistent with the findings in the literature that the proportion of high-

13 High level of education includes ISCED 5 and 6 levels; medium level of education comprises ISCED 3 and 4 levels; and low level of education takes in ISCED 0, 1 and 2 levels. For further details about this classification see UNESCO (1997). 14 Given the construction of the sample, were the propensity to stay in the host country positively correlated with a migrant’s educational attainment (Hazans (2012) shows this to be the case for the Baltic states before enlargement as well as since 2006), our results would underreport the true improvement in the skill composition of immigrants from the new member states.

19

but also low-educated migrants from the EU10 in several EU15 countries increased after

the 2004 enlargement.15

[Figure 8 about here]

As concerns the effects of the 2007 EU enlargement on EU2 migrants in the EU15, we

observe a steady share of high-educated and an increasing share of low-educated

migrants among them during the initial period 2007-2008, followed by a steep increase in

the share of high-educated migrants and a similarly sharp decrease in the share of low-

educated ones in 2009. Whether this signifies a reversal of the trend of decreasing share

of high-educated and an increasing share of low-educated EU2 migrants in the EU

observed during 2001-2007 and to what extent this is caused by Romania’s and

Bulgaria’s EU accession remains to be seen when more recent data become available.

In 2009 among EU10 nationals in the EU15 the share of high educated was 26.1% and

low educated 22.5%; i.e. they were considerably more educated than EU2 nationals in the

EU15 of whom 12.2% were high and 37.5% low educated. They were more educated

than the total population in the EU15 with 18.9% high and 45.7% low educated residents.

EU10 as well as EU2 nationals in the EU15 were each positively selected compared to

their source populations, with 14.4% high educated and 27.4% low educated residents in

EU10 and 10.3% high educated and 40.9% low educated residents in EU2. Most of these

results stay valid if we look at prime working age population (25-54), except that EU2

migrants then appear to be negatively selected form their source population. 15 See Kahanec and Zimmermann (2010).

20

4. The effects of post-enlargement migration in receiving and sending countries

To evaluate the effects of post-enlargement migration in an enlarged EU we consider the

welfare of three key stakeholders to this process: the sending countries, the receiving

countries, and the migrants themselves. Migrants from the new member states in the

EU15 appear to be overrepresented in low- and medium-skilled occupations and sectors,

such as construction, manufacturing, hotels and restaurants, and agriculture (Kahanec,

Zaiceva and Zimmermann, 2010). Given their relatively favorable skill-composition

discussed above, this discrepancy signifies a degree of downskilling and possibly brain

waste. Accompanied with the separation from their families and relatives in their

countries of origin, it is not too surprising that this leads to lack of satisfaction with their

migration experience (Anderson et al. 2006; Blanchflower and Lawton, 2010).

In spite of their possible dissatisfaction with some aspects of their experience as migrants,

post-enlargement migrants can hardly be considered elsewhere but among the winners of

free labor mobility in the EU. Given the wage and unemployment gaps between sending

and receiving countries, post-enlargement migrants have benefited in terms of higher

salaries, improved career prospects, and a generally higher standard of living in the

EU15. Improved human capital and language skills in particular add to the benefits of

their migration experience in the EU15. Kureková (2011) finds that potential employers

value migrant’s work experience acquired abroad upon their return, especially if they are

young. By the revealed preferences argument, the sum of these benefits should exceed

the pecuniary, but also psychological and social, costs migration typically entails.

21

As concerns the effects on receiving countries, the available empirical evidence paints a

rather positive picture. Very small if any effects of post-enlargement migration on the

unemployment rate or wages are found in the UK (Gilpin et al., 2006; Blanchflower

Saleheen, and Shadforth, 2007; Lemos and Portes, 2008). Blanchflower and Lawton

(2010) detect small effects in the least skilled sectors. Blanchflower and Shadforth (2009)

and Blanchflower, Saleheen, and Shadforth (2007) point at the importance of

immigration and the resulting fear of unemployment for suppressing inflationary

pressures. Doyle, Hughes, and Wadensjö (2006) and Hughes (2007) report a similar

picture for Ireland, where post-enlargement immigration might have caused some

substitution and a temporary slow-down of wage growth in some sectors, but any

displacement at the micro level was not affecting aggregate unemployment and the

effects on wage growth reversed soon.

Brenke, Yuksel, and Zimmermann (2010) find that EU8 migrants compete with

immigrants from outside of Europe for low-skilled jobs rather than with the natives. This

may have been one of the causes behind the 50% drop in immigration from other

important source countries, including Russia, Ukraine and Turkey, from 2004 to 2006

reported by these authors. Barrett (2010) argues that post-enlargement immigration

helped Ireland to moderate the rather high wage growth during the pre-2008 boom, which

helped the country in terms of GNP growth. Kahanec and Zimmermann (2009) show that

high-skilled immigration can be expected to decrease inequality, which highlights the

importance of adjustment of high skilled migrants into corresponding jobs. As concerns

the feared effects on the receiving countries’ welfare systems, they have been shown to

22

be unjustified (Gerdes and Wadensjö, 2010; Doyle, 2007; Hughes, 2007). Giulietti et al.

(2012) reject the welfare magnet hypothesis for migration within and into the EU.

The massive outflow of workers from some of the EU10 and EU2 countries has sparked

some fears that the risks of EU enlargement may actually be borne by the new member

states. Kadziauskas (2007) warns that on the background of adverse demographic trends,

the Lithuanian social security system may collapse due to post-enlargement out-

migration. Kaczmarczyk and Okólski (2008) and Kadziauskas (2007) report growing

shortages in some segments of the labor market soon after Poland’s and Lithuania’s EU

accession. Kureková (2011) reports significant skill shortages in Slovakia in the post-

enlargement period. A new trend in the sending countries has emerged, whereby such

skill shortages are filled in by immigrants from outside the EU, mainly from Ukraine,

Belarus, Russia, and some Balkan countries (Frelak and Kazmierkiewicz, 2007; Iglicka ,

2005; Kureková, 2011). Kaminska and Kahancová (2011) report that in Slovakia post-

enlargement outmigration enabled trade unions to obtain wage increases.

An important consideration for the sending countries is to what extent post-enlargement

out-migration represents a lasting loss of labor and human capital and to what extent it

might signify the beginning of an era of brain gain and circulation. Early studies suggest

that there were no signs of significant brain drain, although some skilled sectors, such as

medical doctors, lost non-negligible proportions of their workforce (Frelak and

Kazmierkiewicz, 2007; Brücker et al. 2009; European Commission, 2008b; Hazans

2012). The negative selection into return migration observed for migrants from the Baltic

23

states more recently (Hazans, 2012) may pose some risks for the growth potential and

sustainability of social security in the sending countries.

Also important is to what extent the gains from migration are transmitted to the left-

behinds in the form of remittances. Kahanec, Zaiceva, and Zimmermann (2010) report an

increasing importance of remittances in a number of sending countries, most significantly

in Bulgaria and Romania, but also the Baltic states. In Romania and Bulgaria remittances

constituted about 5 percent of their GDP in 2007 (Dietz, 2009). Comini and Faes-Cannito

(2010) report that the overall volume of remittances to the EU8 and EU2 declined in

2009 after years of growth, probably due to the worsened economic situation in the host

economies due to the financial crisis. Kaczmarczyk and Okólski (2008) document that

remittances were primarily used for consumption and durable goods during the early

post-enlargement period, but also report that more recently they have been invested in

human capital as well. Remittances thus could partly compensate the sending countries

for the possible brain drain.

In a general equilibrium model Baas, Brücker, and Hauptman (2010) argue that the

aggregate GDP of an enlarged EU can be expected to increase by about 0.2%, or 24

billion Euros, from 2004 to 2007 as a consequence of post-enlargement migration from

the EU8 alone. This implies 28,571 EUR per post-enlargement migrant. European labor

markets seem to absorb these flows quite seamlessly, with wages declining by up to 0.1%

in the EU15 and the unemployment rate increasing by about 0.1 percentage points. For

the sending EU8 countries they predict a decline of unemployment of about 0.4

24

percentage points and an increase in wages by about 0.3%. In the long run, however, they

predict no effects on wages or unemployment in the sending or receiving countries.

Similar effects are predicted by Holland et al. (2011), although given a slightly different

modeling approach, the authors predict lasting effects on real wages.

Constant (2011) summarizes this evidence to conclude that the pre-enlargement fears of

labor market disruptions to be caused by immigrants from the new member states were

by and large unjustified. To the contrary, she maintains the migrants and the receiving as

well as the sending countries gained from increased labor mobility in an enlarged EU.

5. The potential for further post-enlargement migration

To shed light on what migration flows can be expected in the foreseeable future, one can

look at current migration intentions. Drawing on Eurostat (2010), Figure 9 reports the

shares of a country’s population that envisage working in another country. One can see

an interesting pattern across an enlarged EU, whereby the most mobile appear to be

Scandinavians, with more than half of the Danes reporting positive intentions. Next and

very close come the Baltic states in each of which more than a third of the population

answered “yes”. Perhaps somewhat surprisingly, at least in view of their relatively low

out-migration rates following their EU accession, Hungary and Slovenia exhibit higher

shares of people who envisage working outside of their country than Poland or Slovakia,

and all exhibit greater shares than seen in Bulgaria and Romania. Probably the main

explanatory factors behind these figures are the relatively adverse economic situation in

25

Slovenia and Hungary, and the comparatively good economic prospects of Slovakia and

especially Poland in late 2009. The low migration intentions of Bulgarians and

Romanians may have to do with the unfavorable economic prospects in some of their

main destination countries in southern Europe. As concerns which destination countries

are preferred by EU8 workers, according to Eurostat (2010) it is mainly Germany

(25.4%) and the UK (25.3%), followed by Austria (13.5%). Workers from EU2 countries

mainly prefer Italy (17.0%), Spain (14.5%) and Germany (14.5%), but also the UK

(11.5%).16

[Figure 9 about here]

A key question is how concrete the intentions in Figure 9 are. Figure 10 reports the

answers of those respondents who envisage working in a country outside their own

country at some time in the future to the question about when they expect it to happen.

The results indicate that migration intentions are most imminent in the Baltic states as

well as Romania and Bulgaria. The remaining new member states, Slovakia and Poland

do not differ very much from the EU27 average, whereas Slovenia, the Czech Republic,

and Hungary exhibit the lowest imminence of migration intentions.

[Figure 10 about here]

16 Simple averages for the EU8 and EU2 countries of shares of respondents indicating preference for the respective country in parenthesis, Eurostat (2010). Based on spontaneous possibly multiple responses including countries outside the EU. For the sake of comparison, the US as the most preferred non-EU country was indicted by 11.3% of EU8 respondents and 12.0% of EU2 respondents.

26

To answer the question from which new member states one can expect the highest

migration outflows in the foreseeable future, we construct a simple analytical migration

imminence matrix using the data on migration intentions as reported in Figures 9 and 10.

We namely plot in Figure 11 the share of population envisaging work abroad against the

share of those of them who indicate that they expect to work there during the next six

months (Panel a) and, as a robustness check, during the next twelve months (Panel b).17

We then interpret the distance from the origin as a measure a country’s imminent

migration potential. In particular, countries that fall into the south-west quadrant of the

migration imminence matrix can be interpreted to have low imminent out-migration

potential. This includes the Czech Republic but also, somewhat surprisingly, Poland and

Slovakia. This may indicate that the migration potential of these countries had been

already partly exhausted by the end of 2009. The countries that fall into the north-west

quadrant, Romania and Bulgaria, exhibit relatively low shares of people planning to work

abroad. However, for a relatively large share of those planning to work abroad the

indicated plans seem to be rather imminent.

[Figure 11 about here]

Slovenia and Hungary fall into the south-east quadrant with relatively high share of

people envisaging work abroad, but only a relatively small share of them indicated this to

happen during the next six months. With Hungary and Slovenia sharing a weak economic

prospect in late 2009 and up until then relatively low out-migration rates, a possible

17 These shares are normalized on the interval [0,1] to range between the respective minimums and maximums observed in the EU.

27

interpretation is that in these countries larger numbers of people were starting to consider

the option of finding a job abroad, but their plans were relatively recent and not concrete

yet.

The highest imminent migration potential is observed in Lithuania and Latvia in the

north-east quadrant, which in 2009 exhibited a relatively high share of people expecting

to work abroad and for this to happen during the next six months. Estonia is the

borderline case with the largest share of people expecting to work abroad in the future

among the EU8+2 countries, although the share of people expecting this to happen during

the next six months is considerably lower than in Latvia and Lithuania and is close to the

EU8+2 simple average. The high degree of similarity between panels a and b indicates

that these findings are robust within the studied horizon of migration intentions.

Based on the migration imminence matrix we can thus conjecture that following the

survey the Baltic states were going to continue to send relatively large numbers of

workers abroad. Romania and Bulgaria still exhibited significant migration potential, but

perhaps some of it has been exhausted by the end of 2009. Such migration fatigue seems

to be even more evident for Slovakia and Poland. The Czech Republic had not been

sending many migrants abroad, and it appears that its low migration potential was not

going to change soon after 2009. Hungary and Slovenia, however, may be the coming

sources of migrants with a larger share of people considering working abroad, although

still without firm short-run plans in late 2009.

28

To fully grasp the prospects of future migration between the new and old member states,

it is necessary to understand the prospect of return migration as well. Migration intentions

of EU10 migrants are known to be rather transitory. For example, of workers registered

in Worker Registration Scheme in the UK in 2008 62% envisaged staying in the UK for

less than three months, up from 59% in 2007 and 55% in 2006 (Kahanec, Zaiceva and

Zimmermann, 2010). The long-run trends in return migration are yet to be evaluated.

First evidence by Hazans (2012) for the Baltic countries indicates that significant shares

of migrants are indeed returning. Whereas they used to be positively selected from

migrant populations in the period immediately following the 2004 enlargement, after

2006 the share of high-skilled workers among returnees is lower than among emigrant

cohorts they come from (ibid.). Such developments could undermine the prospects for

gainful brain circulation from the perspective of sending countries.

Hazans (2012) further reports that compared to the pre-crisis period, out-migration

intensified in Estonia and even more so in Latvia during the crisis. The worsened

economic conditions disproportionally pushed the less skilled as well as ethnic minorities

to migration, mainly Russian-speakers. The author proposes that the stronger response of

Latvians may have to do with their perception of the crisis as not only of financial but

also of a systemic nature. Indeed, Latvian migration became more long-term oriented

during the crisis. The overall increase can be linked to increased unemployment and

worsened economic prospects. Perhaps the most detrimental effect of the crisis is that

high-skilled workers became underrepresented among returnees, undermining the

prospects of gainful brain circulation for these sending countries.

29

7. Conclusions

Aging, diminishing young cohorts and a lack of innovation potential, and structural

mismatches in the labor market resulting in unemployment and skill shortages at the

same time are some of the most important labor market challenges in the EU. These

challenges have contributed to and are themselves aggravated by the current debt crisis in

the Eurozone. On this backdrop embracing the freedom of movement of workers in an

enlarged EU as a powerful tool to improve allocation of human capital and thus combat

some of these challenges would seem rational. Yet, fear and controversies entangled the

implementation of free labor mobility vis-à-vis the countries that joined the EU in 2004

and 2007.

Painstaking empirical analyses based on theoretical underpinnings and hard data

surveyed in this study tell a straight story, however. The free movement of labor in an

enlarged EU can with little doubt be considered a success story of EU integration and

enlargement. It resulted in substantial relocation of labor that has improved the allocation

of human capital in the EU. These new hands and brains appear to have been absorbed by

the receiving labor markets rather seamlessly. In particular, except for some

downskilling, we do not observe any significant negative effects on (un)employment or

wages in the EU15. Similarly, the hypothesis of welfare tourism has not been

substantiated.

30

The sending countries appear to have been relieved of some currently redundant labor

resulting from skill-mismatches in their labor markets, as well being relieved of the

related fiscal burden. Some new skill shortages have emerged, however. Additionally, the

loss of young and skilled labor may be rather worrying in view of the dismal

demographic trends in most of the new member states, as well as for the sustainability of

their public finances. Of key importance for the sending countries is thus their ability to

benefit from brain gain resulting from brain circulation in an enlarged EU. This includes

a proper policy approach to the issues of return and circular migration and inefficient

downskilling. Remittances partly compensate for the loss of human capital possibly

characterizing the early stages of post-enlargement migration. Migrants themselves, as

well as their families, appear to have traded the benefits of migration against some

pecuniary and non-pecuniary costs to their benefit.

Transitional arrangements seem to have affected not only the direction, but also the

composition of post-enlargement migration flows. A full evaluation of their effects is yet

to come, but the evidence so far is that the countries that delayed liberalizing the access

to their labor market for citizens from the new member states disproportionally lost

skilled and young migrants, who chose more welcoming countries such as Ireland and the

UK. Another possibly negative effect is that transitional arrangements led to spurious

self-employment as a strategy to circumvent them.

The current debt crisis in the EU is a challenge on its own. Although the effects of free

labor mobility in the EU are yet to be fully evaluated, based on the available literature we

31

propose that the freedom of movement in an enlarged EU not only contributes to the

European Project by strengthening the social fabric and improving cohesion in the EU, it

does so also by directly contributing to its economic viability. Namely, it provides for an

improved allocative efficiency of European labor markets, a higher innovation potential,

increased utilization of resources and their higher productivity, and the resulting fiscal

relief, all enabling the EU to thrive economically, socially, and politically in a globalized

world.

32

References

Anderson, B., Ruhs, M., Rogaly, B., and Spencer, S. (2006) Fair Enough? Central and

East European Migrants in Low-Wage Employment in the UK. Oxford: COMPAS.

Baas, T., H. Brücker, and A. Hauptmann (2010), Labor Mobility in the Enlarged EU:

Who Wins, Who Loses?, in M. Kahanec and K. F. Zimmermann (eds.), EU Labor

Markets After Post- Enlargement Migration. Berlin et al.: Springer, 47-70.

Barrett, A. (2010) “EU Enlargement and Ireland’s Labor Market,” in M. Kahanec and

K.F. Zimmermann (eds.), EU Labor Markets After Post-Enlargement Migration.

Berlin et al.: Springer, 143-161.

Bauer, T. and Zimmermann K. F. (1999) Assessment of Possible Migration Pressure and

Its Labor Market Impact following EU Enlargement to Central and Eastern Europe.

A study for the Department of Education and Employment, UK. IZA Research Report

No. 3, July.

Becker, G. S. (1957), The Economics of Discrimination. Chicago.

Blanchflower, D. G. and H. Lawton (2010), “The Impact of the Recent Expansion of the

EU on the UK Labour Market,” in M. Kahanec and K. F. Zimmermann (eds.), EU

Labor Markets After Post-Enlargement Migration. Berlin et al.: Springer, 181-215.

Blanchflower, D. G. and C. Shadforth (2009) “Fear, Unemployment and Migration”, The

Economic Journal 119 (535), F136-F182.

Blanchflower, D. G., J. Saleheen and C. Shadforth. (2007), “The Impact of the Recent

Migration from Eastern Europe on the UK Economy,” IZA Discussion Paper No.

2615, Bonn.

33

Bonin, H., W. Eichhorst, C. Florman, M. O. Hansen, L. Skiöld, J. Stuhler, K. Tatsiramos,

H. Thomasen and K. F. Zimmermann (2008), Geographic Mobility in the European

Union: Optimising its Economic and Social Benefits. IZA Research Report No. 19,

Bonn.

Borjas, G. J. (1987), “Self-Selection and the Earnings of Immigrants”, American

Economic Review 77 (4), 531-553.

Borjas, G. J. (1999), “Immigration and Welfare Magnets”, Journal of Labor Economics

17 (4), 607-637.

Brenke, K., M. Yuksel and K. F. Zimmermann (2010), “EU Enlargement under

Continued Mobility Restrictions: Consequences for the German Labor Market”, in M.

Kahanec and K. F. Zimmermann (eds.), EU Labor Markets After Post-Enlargement

Migration. Berlin et al.; Springer, 111-129.

Brücker, H. and A. Damelang (2009), Labour Mobility within the EU in the Context of

Enlargement and the Functioning of the Transitional Arrangements: Analysis of the

Scale, Direction and Structure of Labour Mobility. Background Report, IAB,

Nürnberg.

Brücker, H. et al. (2009), Labour Mobility within the EU in the Context of Enlargement

and the Functioning of the Transitional Arrangements. Final Report (IAB, CMR,

fRDB, GEP, WIFO, wiiw), Nürnberg.

Brücker, H., G. S. Epstein, B. McCormick, G. Saint-Paul, A. Venturini and K. F.

Zimmermann (2002) “Managing Migration in the European Welfare State,” in T.

Boeri, G. Hanson and B. McCormick (eds.), Immigration Policy and the Welfare

System. Oxford, 1-168.

34

Canoy, M., A. Horvath, A. Hubert, F. Lerais, and M. Sochacki (2010) “Post-Enlargement

Migration and Public Perception in the European Union”, in M. Kahanec and K.F.

Zimmermann (eds.), EU Labor Markets after Post-Enlargement Migration, Berlin,

Heidelberg: Springer, 71-107.

Chiswick, B. R. (1980), An Analysis of the Economic Progress and Impact of

Immigrants, Employment and Training Administration. U.S. Department of Labor.

National Technical Information Service, Washington DC, PB80-200454.

Chiswick, B. R. (1998), “The Economic Consequences of Immigration: Application to

the United States and Japan”, in M. Weiner and T. Hanami (eds.), Temporary

Workers or Future Citizens? Japanese and U.S. Migration Policies. New York, 177-

208.

Chiswick, B. R. (1999), “Are Immigrants Favorably Self-Selected?”, American Economic

Review 89 (2), 181-185.

Chiswick, B. R. and P.W. Miller (2011) Negative and Positive Assimilation, Skill

Transferability, and Linguistic Distance IZA Discussion Paper No. 5420, Institute for

the Study of Labor (IZA), Bonn.

Chiswick, C. U.,B. R. Chiswick and G. Karras (1992), “The Impact of Immigrants on the

Macroeconomy”. Carnegie-Rochester Conference Series on Public Policy, 37 (1),

279-316.

Comini, D. and Faes-Cannito, F. (2010). “Remittances from the EU down for the first

time in 2009, flows to non-EU countries more resilient.” Statistics in Focus, vol. 40,

Eurostat.

35

Constant, A.F. (2011) “Sizing it Up: Labor Migration Lessons of the EU Expansion to

27,” Scribani International Conference Proceedings, Bruylant: Belgium, 2012, 49-77.

Constant A.F. and K.F. Zimmermann (forthcoming) “Migration, ethnicity and economic

integration,” forthcoming in Constant A.F. and K.F. Zimmermann (eds.)

International Handbook on the Economics of Migration, Cheltenham: Edward Elgar

Constant A.F., M. Kahanec and K.F. Zimmermann (2009) “Attitudes towards

immigrants, other integration barriers, and their veracity,” International Journal of

Manpower, Emerald Group Publishing 30(1/2), 5-14,

De Giorgi, G. and M. Pellizzari (2006), “Welfare Migration in Europe and the Cost of a

Harmonised Social Assistance”, Labour Economics, 2009, 16 (4), 353-363

de la Rica, S. (2010) “The Experience of Spain with the Inflows of New Labor

Migrants”, in M. Kahanec and K. F. Zimmermann (eds.), EU Labor Markets After

Post-Enlargement Migration. Berlin et al.; Springer, 131-144.

Delbecq, B. A. and Waldorf, B. S. (2010). Going West in the European Union:Migration

and EU-enlargement. Working Paper 10-4. Purdue University, College of

Agriculture, Department of Agricultural Economics.

Dietz, B. (2009). Migration, remittances and the current economic crisis: Implications

for Central and Eastern Europe, Osteuropa Institut Regensburg. Available at:

http://www.oei-dokumente.de/publikationen/info/info-42.pdf.

Doyle, N., G. Hughes, and E. Wadensjö (2006), Freedom of Movement for Workers from

Central and Eastern Europe – Experiences in Ireland and Sweden. Swedish Institute

for European Policy Studies (SIEPS), Report No. 5, Stockholm.

36

Doyle, N. (2007) “The Effects of Central European Labor Migration on Ireland,” in J.

Smith-Bozek (ed.), Labor Mobility in the European Union: New Members, New

Challenges. Washington DC (Center for European Policy Analysis), 35-59.

Dustmann, C., Casanova, M. Fertig, M., Preston, I., and Schmidt, C. M. (2003) The

Impact of EU Enlargement on Migration Flows. Home Office Online Report 25/03.

European Commission (2008a), The Impact of Free Movement of Workers in the Context

of EU Enlargement, Report on the first phase (1 January 2007 – 31 December 2008)

of the Transitional Arrangements Set out in the 2005 Accession Treaty and as

Requested According to the Transitional Arrangements Set out in the 2003 Accession

Treaty. Communication from the Commission to the European Parliament, the

Council, the European Economic and Social Committee and the Committee of the

Regions. Brussels, 18 November 2008.

European Commission (2008b), Employment in Europe, Chapter 3: Geographical labour

mobility in the context of EU enlargement. Directorate-General for Employment,

Social Affairs and Equal Opportunities. Office for Official Publications of the

European Communities. Luxembourg, October 2008.

Eurostat (2010), Geographical and Labour Market Mobility, Special Eurobarometer 337,

wave 72.5, Eurostat, European Commission.

Holland, D., T. Fic, P. Paluchowski, A. Rincon-Aznar and L. Stokes (2011) Labour

Mobility within the EU: The impact of Enlargement and Transitional Arrangements,

NIESR Discussion Paper No. 379, National Institute of Economic and Social

Research, London.

37

Frelak, J. and P. Kazmierkiewicz (2007), “Labor Mobility: The Case of Poland”, in J.

Smith-Bozek (ed.), Labor Mobility in the European Union: New Members, New

Challenges. Washington DC (Center for European Policy Analysis), 60-79.

Galgóczi B., Leschke J, and A. Watt (eds.) (2012) EU Labour Migration in Troubled

Times: Skills Mismatch, Return and Policy Responses, Aldershot: Ashgate

(forthcoming).

Gerdes C. and E. Wadensjö (2010) “Post-Enlargement Migration and Labor Market

Impact in Sweden,” in M. Kahanec and K. F. Zimmermann (eds.), EU Labor Markets

After Post-Enlargement Migration. Berlin et al.; Springer, 163-179.

Gilpin, N., M. Henty, S. Lemos, J. Portes and C. Bullen (2006), “The Impact of Free

Movement of Workers from Central and Eastern Europe on the UK Labour Market,”

Department of Work and Pensions, Working Paper No. 29, London.

Giulietti, C., M. Guzi, M. Kahanec, and K. F. Zimmermann. 2011. “Unemployment

Benefits and Immigration: Evidence from the EU”, IZA Discussion Paper No. 6075,

Bonn: Institute for the Study of Labor. Forthcoming in International Journal of

Manpower, 2012.

Harris, J.R. and M.P. Todaro (1970), “Migration, Unemployment and Development: A

Two-Sector Analysis”, American Economic Review 60 (1), 126-142.

Hazans, M. (2012) “Selectivity of migrants from Baltic countries before and after

enlargement and responses to the crisis,” in Galgóczi B., Leschke J, A. Watt (eds.)

EU Labour Migration in Troubled Times: Skills Mismatch, Return and Policy

Responses, Aldershot: Ashgate (forthcoming).

38

Hazans, M. and K. Philips (2010) “The Post-Enlargement Migration Experience in the

Baltic Labor Markets,” in M. Kahanec and K. F. Zimmermann (eds.), EU Labor

Markets After Post-Enlargement Migration. Berlin et al.; Springer, 255-304.

Hughes, G. (2007), EU Enlargement and Labour Market Effects of Migration to Ireland

from Southern, Central and Eastern Europe (Paper presented at Second IZA

Migration Workshop: EU Enlargement and the Labour Markets, Bonn, 7-8 September

2007).

Iglicka, K. (2005), The Impact of the EU Enlargement on Migratory Movements in

Poland. Center for International Relations, Report No. 12/05, Warsaw.

IOM (1998) Migration Potential in Central and Eastern Europe. Geneva: International

Organization for Migration.

Kaczmarczyk, P. and M. Okólski (2008), Economic Impacts of Migration on Poland and

the Baltic States. Fafo-Paper 2008: 1, Fafo, Oslo.

Kaczmarczyk P., M. Mioduszewska and A. Żylicz (2010) “Impact of the Post-Accession

Migration on the Polish Labor Market,” in M. Kahanec and K. F. Zimmermann

(eds.), EU Labor Markets After Post-Enlargement Migration. Berlin et al.; Springer,

219-253.

Kadziauskas, G. (2007), “Lithuanian Migration: Causes, Impacts and Policy Guidelines,”

in J. Smith-Bozek (ed.), Labor Mobility in the European Union: New Members, New

Challenges. Washington DC (Center for European Policy Analysis), 80-100.

Kahanec, M. and K. F. Zimmermann (2009), “International Migration, Ethnicity and

Economic Inequality”, in W. Salverda, B. Nolan and T. M. Smeeding (eds.), The

Oxford Handbook of Economic Inequality. Oxford, 455-490.

39

Kahanec, M. and K.F. Zimmermann (eds.) (2010) EU Labor Markets after Post-

enlargement Migration. Berlin: Springer.

Kahanec, M., A. Zaiceva, A., and K.F. Zimmermann (2010) “Lessons from migration

after EU enlargement,” in M. Kahanec and K.F. Zimmermann (eds.), EU Labor

Markets after Post-Enlargement Migration, Berlin, Heidelberg: Springer, 3-45.

Kahanec, M., Anna M.-H. Kim and K. F. Zimmermann (2011), Pitfalls of Immigrant

Inclusion into the European Welfare State, IZA Discussion Paper 6260, Institute for

the Study of Labor (IZA), Bonn.

Kaminska M. E. and M. Kahancová (2011), ‘Emigration and labour shortages: An

opportunity for trade unions in the New Member States?’, European Journal of

Industrial Relations, 17 (2) 189-203.

Kureková, L. (2011) From job search to skill search. Political economy of labor

migration in Central and Eastern Europe. PhD dissertation, Central European

University (CEU), Budapest.

Layard, R., Blanchard, O., Dornbusch, R., and Krugman P. (1992) East–West Migration.

The Alternatives. Cambridge, MA: MIT Press and WIDER.

Massey, D. S. (1990), “Social Structure, Household Strategies, and the Cumulative

Causation of Migration”, Population Index 56 (1), 3-26.

Mincer, J. (1978), “Family Migration Decisions”, Journal of Political Economy 86 (5),

749-773.

Ottaviano, G.I.P., and G. Peri (2006) “The Economic Value of Economic Diversity:

Evidence from US Cities”, Journal of Economic Geography 6, 9-44.

40

Sinn, H.-W. et al. (2000) EU-Erweiterung und Arbeitskräftemigration Wege zu einer

schrittweisen Annäherung der Arbeitsmärkte. Studie im Auftrag des Bun-

desministeriums fur Arbeit und Sozialordnung. München: Ifo Institut.

Sinn, H.-W. and W. Ochel (2003) “Social Union, Convergence and Migration”, Journal

of Common Market Studies 41 (5), 869–896.

Sjaastad, L. A. (1962), “The Costs and Returns of Human Migration”, Journal of

Political Economy 70(5), 80-93.

Stark, O. (1991), The Migration of Labor. Cambridge.

Tijdens, K. and M. van Klaveren. (2012) “A skill mismatch for migrant workers?

Evidence from WageIndicator survey data,” in Galgóczi B., Leschke J, A. Watt (eds.)

EU Labour Migration in Troubled Times: Skills Mismatch, Return and Policy

Responses, Aldershot: Ashgate (forthcoming).

UNESCO (1997) International Standard Classification of Education, Paris: UNESCO.

Zaiceva, A. and K. F. Zimmermann (2008), “Scale, Diversity, and Determinants of

Labour Migration in Europe”, Oxford Review of Economic Policy 24 (3), 427-451.

Zimmermann, K.F. 2005. European Migration: What do we know? Oxford/New York:

Oxford University Press.

Zimmermann, K.F., M. Kahanec, A. Barrett, C. Giulietti, B. Maître and M. Guzi (2012)

“Study on active inclusion of immigrants”, IZA Research Report 43, Institute for the

Study of Labor (IZA), Bonn.

41

Tables and Figures

Table 1. Citizens from new EU member states residing in the EU15.

Migrants from the EU8 Migrants from the EU2

Total Percent of

EU8 population

Percent of EU15

population Total

Percent of EU2

population

Percent of EU15

population

1997 673,324 0.91 0.18 249,781 0.81 0.07 1998 674,972 0.91 0.18 234,743 0.76 0.06 1999 717,976 0.97 0.19 271,657 0.88 0.07 2000 753,056 1.02 0.20 315,699 1.03 0.08 2001 800,534 1.09 0.21 391,045 1.28 0.10 2002 851,250 1.16 0.22 509,160 1.71 0.13 2003 942,321 1.29 0.25 711,930 2.40 0.19 2004 1,006,851 1.38 0.26 916,298 3.10 0.24 2005 1,235,429 1.69 0.32 1,109,570 3.77 0.29 2006 1,627,625 2.23 0.42 1,376,956 4.69 0.35 2007 2,027,651 2.78 0.52 1,971,968 6.74 0.50 2008 2,252,681 3.09 0.57 2,348,523 8.05 0.60 2009 2,288,600 3.13 0.58 2,564,008 8.81 0.65

Source: Based on data provided in Holland et al. (2011), Eurostat population statistics, and own calculations.

42

Figure 1. Effects of skilled and unskilled immigration

Figure 2. Effects of skilled and unskilled out-migration

1UL

Immigration

Labor

Wage Wage

Labor

A0

A1

A2

(a) High-skilled (b) Low-skilled

0L

0hw 2hw 1hw

1lw0lw

0UL

0UB 1

UB1B0B

1hD

0hD

1H0H

1lD

0lD

Immigration

2lw

Ulw

1L

3hw

2B

3A

Out-migration Wage Wage

Labor

M0

M1

M2

(a) High-skilled (b) Low-skilled

0L

0hw

2hw

1hw

1lw

0lw

0UN1

UN

1N

0N1hD

0hD

1H 0H

1lD

0lD

Out-migration

2lw

Ulw

1L

3hw

2N

3M

0UL1

UL

43

Figure 3. Net inflows of EU8 and EU2 citizens to the EU15.

-100,000

0

100,000

200,000

300,000

400,000

500,000

600,000

700,000

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

EU8

EU2

Source: Own calculations based on Holland et al. (2011) data.

Figure 4. The share of sending country’s population residing in the EU15. a. Visegrad countries and Slovenia b. The Baltic states

0

1

2

3

4

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

CZ

HU

PL

SK

SI

0

1

2

3

4

5

6

7

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

EE

LV

LT

c. Countries acceding the EU in 2007 d. EU8 and EU2

0

2

4

6

8

10

12

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

BG

RO

0

1

2

3

4

5

6

7

8

9

10

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

EU8

EU2

Notes: In percent of source country’s population. Source: Own calculation based on Holland et al. (2011) data and Eurostat population statistics.

44

Figure 5. The share of EU8 citizens residing in EU15. a. Liberalization as of May 1, 2004 b. Liberalization in May – July 2006

0

1

2

3

4

5

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Ireland

Sweden

UK

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Greece

Spain

Italy

Portugal

Finalnd

c. Liberalization May 2007 – May 2009 d. Liberalization as of May 1, 2011

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Netherlands

Luxembourg

France

Belgium

Denmark

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Germany

Austria

Notes: In percent of host country’s population. Source: Own calculation based on Holland et al. (2011) data and Eurostat population statistics. Figure 6. The share of EU2 citizens residing in the EU15. a. Southern Europe b. Scandinavia

0.0

0.4

0.8

1.2

1.6

2.0

2.4

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Greece

Spain

Italy

Portugal

0.00

0.03

0.05

0.08

0.10

0.13

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Finland

Denmark

Sweden

c. Ireland and the UK d. Western Europe

0.0

0.1

0.2

0.3

0.4

0.5

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

UK

Ireland

0.0

0.2

0.4

0.6

0.8

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Luxembourg

Austria

Germany

Belgium

Netherlands

France

Notes: In percent of host country’s population. Source: Own calculation based on Holland et al. (2011) data and Eurostat population statistics.

45

Figure 7. Proportions of EU8 and EU2 citizens in EU15. a. EU8

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

UK

Sweden

Finland

Portugal

Austria

Netherlands

Luxembourg

Italy

France

Spain

Greece

Ireland

Germany

Denmark

Belgium

b. EU2

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

UK

Sweden

Finland

Portugal

Austria

Netherlands

Luxembourg

Italy

France

Spain

Greece

Ireland

Germany

Denmark

Belgium

Notes: In percent of total EU8 and EU2 populations resident in EU15. Source: Own calculation based on Fic 2011 data and Eurostat population statistics.

46

Figure 8. Educational attainment of EU10 and EU2 citizens in the EU15. a. EU10

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Total

low

medium

high

b. EU2

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Total

low

medium

high

Source: Own calculation based on the EU Labour Force Survey, 2009. Notes: In percent of total EU8 and EU2 populations resident in the EU15 above and including 16 years of age, excluding conscripts on compulsory military or community service as well as anyone whose highest level of education or training successfully completed was attained after his or her immigration to the current country of residence in the EU15. No data available for Malta. Germany excluded due to no information on migrants’ country of birth.

47

Figure 9. Migration intentions in Europe

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%D

K

EE

SE LV LT FI

SI

HU

EU

8+2

UK

FR PL

SK IE NL

PT

MT

BE

EU

27 BG LU RO ES

CZ

DE

CY

EL

AT IT

DK

No

Yes

Source: Eurostat (2010), data collection November-December 2009. Notes: Answers to the question QC10: Do you envisage to work in a country outside (our country) at some time in the future? DK stands for “do not know”, or no answer. Sorted by “Yes”. EU8+2 calculated as a simple average of the values for the EU8 and EU2 countries.

Figure 10. The time horizon of migration intentions in Europe

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

LV RO

BG LT LU PT

EU

8+2

EE

MT

FR IE SE

EU

27 PL

ES

SK EL

NL

UK

DE IT BE

DK AT SI

FI

CZ

HU

CY

DK

DK yet

>5 years

3-5 years

1-2 years

6-11 months

<6 months

Source: Eurostat (2010), data collection November-December 2009. Notes: Answers to the question QC11: How soon are you likely to work there? The universe consists of those respondents that answered “Yes” to QC10: Do you envisage to work in a country outside (our country) at some time in the future? “DK yet” stands for “do not know yet; DK stands for “do not know”, or no answer. Sorted by <6 months. EU8+2 calculated as a simple average of the values for the EU8 and EU2 countries.

48

Figure 11. The migration imminence matrix, EU8+2 a. Work abroad within six months b. Work abroad within a year

RO

LT

HUCZ

BG

SKSI

PL

LV

EU27

EE

EU8+2

0

0.5

1

0 0.5 1

Normalized share envisaging to work abroad

No

rma

lize

d s

ha

re e

nvi

sag

ing

to w

ork

ab

roa

d w

ithin

the

ne

xt 6

mo

nth

s

RO

LT

HUCZ

BG

SK SI

PL

LV

EU27

EEEU8+2

0

0.5

1

0 0.5 1

Normalized share envisaging to work abroad

No

rma

lize

d sh

are

en

visa

gin

g to

wo

rka

bro

ad

with

in th

e n

ext 1

2 m

onth

s

Source: Eurostat (2010), data collection November-December 2009. Notes: Based on the answers to the question QC10: “Do you envisage to work in a country outside (our country) at some time in the future?” and QC11: “How soon are you likely to work there?”. The share of population answering “Yes” to the first question is on the x-axis, and, of those, the share answering “During the next 6 months” is on the y-axis. Values normalized with 1 representing the highest, and 0 the lowest, value observed in the EU27. EU8+2 calculated as a simple average of EU8+2 countries.